

This file outlines the items available in our replication materials.

* pretest.csv, pretest.RData
- 32 columns for 2,791 respondents
- The .csv and .RData files are identical
- These data are used only in Section C of the online appendix, as well as any references to that section of the appendix in the paper itself
- These data are a pilot test of using university students of the outgroup antipathy measure, comparing it to other common measures of outgroup sentiment

* stud01_deID.csv, stud01_deID.RData
- 109 columns for 5,811 respondents
- The .csv and .RData files are identical
- These are the data from study 1 and are used in all study 1 analyses in the paper and the online appendix
- These data come from three samples: voters, citizen activists, and elected officials. More information on this data can be found in Section A.1 of the online appendix.

* stud02_deID.csv, stud02_deID.RData
- 70 columns for 2,159 respondents
- The .csv and .RData files are identical
- These are the data from study 2 and are used in all study 2 analyses in the paper and the online appendix
- These data come from a two-wave survey of statewide voters obtained from exit polls. More information on this data can be found in Section A.2 of the online appendix.
- Note that these files are a subset of the total dataset, downloaded at the closing of study 2. This dataset, when analyzed in our replication code, matches the final row of Figure A.2 in the online appendix. For a full accounting of how we reached the numbers in that flowchart, see the script "appendix_flowchart_study2.R" and associated data files, as described below.

* ChangingHearts_GublerEtAl_Jul2021.R
- An R script that produces all figures, tables, and analyses found in the paper and the online appendix.

* ChangingHearts_GublerEtAl_Jul2021_log.pdf
- A log of the output of this R script when it is run by the researchers.

* appendix_flowchart_study2.R, stud02_pre.csv, stud02_post.csv
- The "pre" script includes 2 columns for 3,624 respondents
- The "post" script includes 11 columns for 2,632 respondents
- These datasets are a subset of the wave 1 and wave 2 surveys for study 2. Unlike the data provided for the main replication script, the pre- and post- datasets here include respondents who responded to the survey after the study closed and, therefore, were not included in our analysis in the paper and the appendix. We provide these data to allow the replication of the numbers in the flowchart in Figure A.2 in the online appendix.
- The R file contains the minimal code necessary to replicate that flowchart
